As mentioned in the previous post, a system is always of something. It may be biological, social, or computer-based, all can be understood as systems even machines are systems, just usually not complex ones. We will find ourselves facing a challenge from a specific type of system, perhaps transportation or perhaps environmental. Each challenge will have a specific occurrence. The specifics of an environmental challenge in Florida will not be the same as those in North Dakota. The specifics of how such individual systems work, how the elements making up those systems actually interact has not really been delved into to any extent. How do you begin to understand the basic dynamic interactions of systems? The most fundamental means of doing so is building models of those systems.
A model is an illustration of the interrelationship between parts or elements of a system that explains how they are related and that can also help in understanding why. A model can be anything from a 3D computer simulation to a paper napkin drawing. So the idea then is to make models.
Is there then some tool that anyone could use to build and experiment with models to test out some of the general ideas of systems thinking and apply them to specific challenges? It would need to be relatively easy to use though capable of doing more complex actions as you learn. It should also be web-based and on the ‘cloud’ rather than be a stand alone desktop program. This would help in at least three ways. One, it would make it more available. Two, there could be a repository of different models and three, it could be possible to let people copy or clone models that they found interesting. Finally, in the best of all possible worlds, it would be free.
There is good news, a little bad and more good news. First, there is such a program called Insight Maker or Insightmaker.com, which does everything listed above. The little bit of bad news is that this current course is only going to give you the bare basics, though it is a good start. The final good news is that they are planning a course focused on Insight Maker in September of 2014.
Insight Maker has already been introduced through the previous articles on this course and was the means of making the models used so far. It has also been featured on this blog prior to the course. However, because those earlier posts dealt with other issues they may not have been the best means of introducing Insight Maker.
Once again using a Kumu map, you start at the green circle and move along a path, first to qualitative models which include Rich Pictures and Causal Loop Diagrams and then quantitative models with a brief visit to Stock and Flow diagrams prior to reaching a point at which you are introduced to Your First Simulation and then finishing at the red colored end circle that summarizes the segment. In general terms, the segment seeks to cover the following concepts.
• Model. A simplification of reality intended to promote understanding
• Qualitative Models. Static models depicting the relationships between elements of the model. Rich Pictures and Causal Loop Diagrams are two examples of qualitative models.
• Quantitative Models. Dynamic models allow one to experience the implications of the relationships of the elements of the model over time.
◦ Testing every element added to a dynamic model will serve you well.
• Model Development. This tends to be a recursive learning process migrating toward understanding.
• Comments. While it may feel unnecessary, or even an unnatural act, adding comments which embrace your thoughts during the model development will serve you well in the future. You will be surprised how much you forget about the thoughts behind your model in a week, a month, or a year.
The nuts and bolts of the segment though are on using Insightmaker.com. Insight Maker was designed as a simulation environment though may be used to do multiple types of models. Which type is most appropriate depends on the situation to be addressed.
It is better to jump right into Insightmaker.com and play around with it. With some basic instruction, you can start making simple models and test them out. To be fair, I also have Beyond Connecting the Dots, the book written by Gene Bellinger, the instructor for the course and Scott Fortmann-Roe, the creator of Insight Maker. The course material does though give you the basics. You just have to throw yourself into doing some trial and learning. Based on the fundamental premise that we are creating models to get a greater and deeper understanding, the common concept of trial and error never applies, it is always what Gene Bellinger calls trial and learning. You are not going to break anything or cause any systems failure simply by making models with Insight Maker and it can develop that sought after deeper understanding that systems thinking calls for.
Now I am going to say something about modeling that, although it has the potential of driving away people who crave certainty, always needs to be kept in mind. It is a fundamental premise for experienced systems thinkers, pronounced by one of the field’s major thinkers.
All models are wrong but some models are useful - George Box.
The assigned work for this segment called for creating different models. A Rich Picture model and a Causal Loop Diagram in story format were made before. What had not been attempted so far is a Stock and Flow diagram
Following the jumping in and swimming plan of action, the course’s Your First Simulation (IM-13312), which has basic model building blocks, was cloned then copies made (using command-select, command-c and command-v on a Mac) of the three examples provided, which were expanded upon creating Simulation Model Examples (Clone from Your First Simulation Model) then using those same building blocks, stocks, flow and variables to get a better understanding of how they worked.
Stocks represent the ‘stuff’ that can be measured or counted, the inventory of products or census of a population. Stock can also be increased or decreased but not instantaneously. In some cases, it may change extremely quickly but this usually occurs over extended time. Flow is the means of increasing, by flowing in and decreasing by flowing out from stock. Variables influence the rate at which flow occurs by varying the stock or the flow in some manner. Together, these can be constructed into a graphic mathematical formula that serves as the engine for the model. To get a better understanding of how the models (and modeling) were working, I also re-created a particular model's formulas in a spreadsheet format using Numbers.
With this knowledge, my first Stock and Flow diagram was finally finished. Well, completed but not finalized as I still have other ideas. There was a good deal of trial and learning involved, more than I expected, with numerous redesigns and a few restarts. There is the issue of not only making sure your model is actually modeling the real world circumstances that you are considering to the extent needed but also making sure that the internal workings of your model are also consistent with that endeavor. In early versions, the external numbers looked good but a closer examination indicated that it was not calculating what I thought it was. I checked by putting the numbers into a spreadsheet to see if the presumed equations generated by the model could be regenerated there.
Below the model’s description are sliders that you can use to change the number of homes, businesses and other traffic. Pressing the ‘Run Simulation’ button at the top of the page shows the effects based on the numbers over a 20 year time period. My model ostensibly calculates road wear and tear on community roads due to increased traffic particularly through new development. What it actually considers is the relation between elements within the modeled system.
My model is wrong, remember George Box. It only examines a slice of the world, ignoring other aspects. Its knowledge of the world is not accurate and what it does examine is not realistically portrayed with any precision. I tried to make its limitations clear in the model. Hopefully, though it could be useful, or be the basis for something, if only as a practice run, for something that could become useful.
In writing about the real world understanding mistakes made by Watson, the IBM computer that defeated the greatest champions of the television quiz show Jeopardy at their own game, to explain the difference between knowledge and understanding. Michael Strevens wrote:
“Watson and you both answer questions by seeing connections between things. But they are different kinds of connections. Watson picks up from things it reads that there is a correlation between a sphere’s rotating and a fixed point on its surface having a constantly changing view of the rest of the world. You grasp why this correlation exists, seeing the connection between the opacity of the Earth, light’s traveling in straight lines, and geometry of the sphere itself. For you the statistics are a byproduct of what really matters, the physical and causal relations between things and people and what they do and say. Grasping those relations is what understanding consists in. Watson lives in a world where there are no such relations: all it sees are statistics. It can predict a lot and so it can know a lot, but what it never grasps is why its predictions come true.”
Systems thinking and modeling can help you to understand why.